About half of the infants congenitally infected with HIV-1 seem to acquire the virus intrapartum. Mucosal exposure during birth is likely to be an important route of infection. Our initial longterm goal had been to develop immunoprophylaxis to prevent intrapartum HIV-1 transmission. We established a primate model for mucosal infection at birth and demonstrated that SIV-delta3 (nef-, vpr-, NRE-), which acted as live attenuated virus vaccine in adult macaques, could infect rhesus monkey fetuses and neonates. However, all macaque neonates given this virus orally at birth thus far developed signs of AIDS, and half have died of disease. PCR analysis showed that the deletions in nef had not been repaired, and that SIV-delta3 had undergone further deletions after in vivo passage in some macaque neonates. These surprise findings indicate that the multiply deleted SIV mutant has retained its pathogenic potential, and that neither nef nor vpr are required for the development of AIDS. The overall goal of this (adjusted) proposal is to examine the molecular determinants for pathogenicity and avirulence in congenital SIV infection. The Specific Aims are to: 1. Continue to evaluate the incidence and spectrum of disease following congenital SIV-delta3 infection. Animals, infected at various time points during gestation or at birth, will be followed longterm. 2. Clone virus directly from tissues of diseased offspring originally infected with SIV-delta3 and determine the DNA sequence. A replication- competent virus will be generated and tested for pathogenicity or attenuation in macaque mother/infant pairs. 3. Test whether the pathogenicity of SIV-delta3 is correlated with the initial inoculum given. 4. Clone virus directly from tissues of offspring that did not develop disease and sequence. Cloned virus from healthy, infected offspring will be tested in adults and neonates to test whether it remained avirulent. In case all offspring exposed to SIV-delta3 develop disease, other candidate attenuated SIVs will be tested in mucosally infected neonatal macaques. 5. Construct virus recombinants and search for the molecular determinants of pathogenicity. The viruses will be tested in mucosally infected neonatal macaques, which our data have shown to be sensitive indicators of the potential pathogenicity of mutant SIVs. The significance of this proposal lies in the systematic evaluation of the lentiviral genes that determine the development of AIDS.